3 Replies Latest reply on Jan 4, 2010 8:00 PM by frankas

    SDKv2.0  Catalyst9.12h (linux) leaves primary display blank

    frankas

      I tried upgrading driver for 5850 with Catalyst_9.12_Hotfix_Linux_8.682.2RC1_Dec15.zip ( I assume this is the driver I need for 2.0 SDK ) But when I add the second Screen (card) to ServerLayout section in xorgs.conf - X wont start, but hangs on a black screen with signal, but no picture. Not even the annoying "Testing use" watermark. Removing the second screen works fine, but only a single GPU is avialable. When X hangs on a black screen, it is possible to use both GPUs from a text consol I can even connect through x11vnc and averything seems to work normally until I move the mouse, at which point vnc only shows screen updates, but doesn't recieve more input.. X will also respond to CTRL - backspace when "hanging" This seems to be a regression from the 9.10 driver (8.661) where this setup actually worked.

      My setup is:

      Ubunty (jaunty) 2.6.28-17-generic

      2 x HD5850 without CrossfireX connected

      1 Monitor conected to the first card. (no monitors attached to the second)

      My xorg.conf is attached. I don't need a second screen, just 2 GPUs, and preferably no watermark. Is there a newer driver I can use with SDK v2.0 - or do I have to revert to the old (fglrx_8.661) ? Please help...

       

      Section "ServerLayout" Identifier "aticonfig Layout" Screen 0 "aticonfig-Screen[0]-0" 0 0 ####### COMMENTING OUT THIS CAUSES X TO HANG ########## # Screen "aticonfig-Screen[1]-0" RightOf "aticonfig-Screen[0]-0" EndSection Section "Files" EndSection Section "Module" EndSection Section "Monitor" Identifier "aticonfig-Monitor[0]-0" Option "VendorName" "ATI Proprietary Driver" Option "ModelName" "Generic Autodetecting Monitor" Option "DPMS" "true" EndSection Section "Monitor" Identifier "aticonfig-Monitor[1]-0" Option "VendorName" "ATI Proprietary Driver" Option "ModelName" "Generic Autodetecting Monitor" Option "DPMS" "true" EndSection Section "Device" Identifier "aticonfig-Device[0]-0" Driver "fglrx" Option "DRI" "true" BusID "PCI:1:0:0" EndSection Section "Device" Identifier "aticonfig-Device[1]-0" Driver "fglrx" # Option "DRI" "true" BusID "PCI:2:0:0" EndSection Section "Screen" Identifier "aticonfig-Screen[0]-0" Device "aticonfig-Device[0]-0" Monitor "aticonfig-Monitor[0]-0" DefaultDepth 24 SubSection "Display" Viewport 0 0 Depth 24 EndSubSection EndSection Section "Screen" Identifier "aticonfig-Screen[1]-0" Device "aticonfig-Device[1]-0" Monitor "aticonfig-Monitor[1]-0" DefaultDepth 24 SubSection "Display" Viewport 0 0 Depth 24 EndSubSection EndSection Section "DRI" Mode 0666 EndSection

        • [Deleted]
          frankas

           

           

           

          • SDKv2.0  Catalyst (linux) leaves primary display blank
            frankas

            I did some more debugging on this issue, and attached a monitor to the secondary card. To my surprise the secondary display works fine.

            I have tried juggling the displays around in xorg.conf, and the result is always the same: Primary display is left blank, while the secondary works. Unfortunately I haven't been able to get a setup I can work with, since most windows will open on the primary display, which I can only reach with VNC.

            Similar problem is reported here:

            http://forums.amd.com/game/messageview.cfm?catid=264&threadid=124248

            It seems that this a regression after 9.10 which did work (for me), as such it is purely a driver bug, and should be reported elsewhere. But where should I report it ?  tried using the feedback in the gaming section, but I couldn't report Linux bugs there.

             

             

             

              • SDKv2.0  Catalyst (linux) leaves primary display blank
                frankas

                I connected the crossfire, but I still cant get a single display with 2 GPUs. However I did catch this in dmesg:

                [  816.847141] Assertion failed in ../../../../../../../../drivers/2d/lnx/fgl/drm/kernel/hal_evergreen.c at line: 63
                [  816.866493] Assertion failed in ../../../../../../../../drivers/2d/lnx/fgl/drm/kernel/hal_evergreen.c at line: 63

                Ahhh... the wonders of closed source drivers...